Copilot Workspace: Auto-validation, go to definition, and more

Copilot Workspace header

We’re excited to share the latest improvements for Copilot Workspace, including enabling enterprise managed users, auto-validating changes in the workspace, and more. Let’s get started! 🚀

Auto-validation

We’ve shipped a new experiment enabling Copilot Workspace to automatically perform a build & test after implementation. If any failures are detected, it will attempt repairs.

To enable this experimental feature, go to Experiments > Start verify loop after implement.

Go to definition

We’ve enabled go to definition support in the editor to improve code navigation.

Photo showing Go-to-Definition

File-specific plan

While you’re implementing your plan with Copilot Workspace, you will now be able to view and edit the file-specific plan items. These are located right above the file, so you can seamlessly edit and iterate on your plan.

File tree reflects plan changes

While iterating on your plan, when you select or deselect a particular plan item, these changes will now be reflected in the file tree in real-time.

Support for enterprise managed users

GitHub Enterprise Cloud customers with Enterprise Managed Users can now use Copilot Workspace. Enterprise administrators have full control over enabling Workspace in their enterprise. To do so:

  1. Enable the Copilot Workspace policy through your enterprise account (note this means accepting the GitHub Next Terms and Conditions)
  2. Enable Preview features and Copilot Extensions at the enterprise or organization level. Please note that if any organization a user belongs to has either of these policies disabled, then they are considered disabled for that user throughout the site, regardless of context. The user will not have access to Copilot Workspace or other features that require those policies.
  3. Ensure the organization has valid Copilot access for users of Copilot Workspace.
  4. Enable the GitHub Next OAuth Application in the organization.

As we continue to evolve Copilot Workspace, we’ll be simplifying these steps, so stay tuned for improvements!

Edit and validate Autofix suggestions

GitHub Advanced Security users can now use Copilot Workspace for pull requests to triage and apply Copilot Autofix suggestions, with the ability to build, test and run the proposed changes without leaving GitHub. Learn more.

Bug fixes & other improvements

  • Added a notification so users know when they need to update the spec after the task has changed.
  • Changed search so when a file is selected, it is highlighted in the file tree.
  • Enabled renaming files on the file page, rather than just on the files changed page.
  • Changed plan to auto-hide when there is no plan defined.

Providing feedback

Please give feedback in our GitHub Discussion. Any and all feedback is appreciated!

Edit and validate Copilot Autofix suggestions with Copilot Workspace

Copilot Autofix suggestions for code scanning alerts can now be edited and validated using Copilot Workspace for pull requests.

Copilot Workspace for Copilot Autofix for code scanning

With this, GitHub Advanced Security users can:

  • Review and integrate Copilot Autofix suggestions within the context of the pull request, benefiting from an improved diff-viewing experience.
  • Refine and address code scanning alerts directly within the pull request, utilizing an enhanced code editing experience.
  • Build, test, and run proposed changes in the pull request without impacting your personal build and test environment.

All GitHub Advanced Security users can use this feature in private repositories on GitHub.com. A Copilot license is not required.

To learn more about code scanning alerts and Copilot Autofix, see About Copilot Autofix for CodeQL code scanning. If you have feedback regarding Copilot Autofix for code scanning, please join the discussion here.

See more

OpenAI o3-mini now available in GitHub Copilot and GitHub Models (Public Preview)

Copilot and GitHub Models o3-mini Release

OpenAI’s latest model, o3-mini, is now available in GitHub Copilot and GitHub Models, bringing OpenAI’s newest reasoning model to your coding workflow.

The o3-mini reasoning model outperforms o1 on coding benchmarks with response times that are comparable to o1-mini, meaning you’ll get improved quality at nearly the same latency.

This cutting-edge model is rolling out gradually and will be available to GitHub Copilot Pro, Business, and Enterprise users today via the model picker in Visual Studio Code and github.com chat (support in Visual Studio, and JetBrains are coming soon). To accelerate your workflow, whether you’re debugging, refactoring, modernizing, testing, or just getting started, simply select “o3-mini (Preview)” to begin using it.

Paid Copilot subscribers get up to 50 messages every 12 hours. Business or Enterprise admins can enable o3-mini access for org members through their org and enterprise admin settings pages.

GitHub Models users with a paid Copilot plan will also be able to leverage the o3-mini model to enhance their AI applications and projects later today. In the GitHub Models playground, you can explore o3-mini’s versatility as you experiment with sample prompts, refine your ideas, and iterate as you build. You can also try it alongside other models available on GitHub Models including models from Cohere, DeepSeek, Meta, and Mistral.

To learn more, check out product documentation on GitHub Models. You can also join our community discussions.

See more
View all changes